Early Perihematomal Edema Expansion: Definition, Significance, and Association with Outcomes after Intracerebral Hemorrhage

Hybrid clinical-radiomics model based on fully automatic segmentation for predicting the early expansion of spontaneous intracerebral hemorrhage: A multi-center study

BACKGROUND

Early prediction of hematoma expansion (HE) is important for the development of therapeutic strategies for spontaneous intracerebral hemorrhage (sICH). Radiomics can help to predict early hematoma expansion in intracerebral hemorrhage. However, complex image processing procedures, especially hematoma segmentation, are time-consuming and dependent on assessor experience. We provide a fully automated hematoma segmentation method, and construct a hybrid predictive model for risk stratification of hematoma expansion.

PURPOSE

To propose an automatic approach for predicting early hemorrhage expansion after spontaneous intracerebral hemorrhage using deep-learning and radiomics methods.

METHODS

A total of 258 patients with sICH were retrospectively enrolled for model construction and internal validation, while another two cohorts (n=87, 149) were employed for independent validation. For hemorrhage segmentation, an iterative segmentation procedure was performed to delineate the area using an nnU-Net framework. Radiomics models of intra-hemorrhage and multiscale peri-hemorrhage were established and evaluated, and the best discriminative-scale peri-hemorrhage radiomics model was selected for further analysis. Combining clinical factors and intra- and peri-hemorrhage radiomics signatures, a hybrid nomogram was constructed for the early HE prediction using multivariate logistic regression. For model validation, the receiver operating characteristic (ROC) curve analyses and DeLong test were used to evaluate the performances of the constructed models, and the calibration curve and decision curve analysis were performed for clinical application.

RESULTS

Our iterative auto-segmentation model showed satisfactory results for hematoma segmentation in all four cohorts. The Dice similarity coefficient of this hematoma segmentation model reached 0.90, showing an expert-level accuracy in hematoma segmentation. The consumed time of the efficient delineation was significantly decreased, from 18 min to less than 2 min, with the assistance of the auto-segmentation model. The radiomics model of 2-mm peri-hemorrhage had a preferable area under ROC curve (AUC) of 0.840 (95 % confidence interval [CI]: 0.768, 0.912) compared with the original (0-mm dilatation) model with an AUC of 0.796 (95 % CI: 0.717, 0.875). The clinical-radiomics hybrid model showed better performances for HE prediction, with AUC of 0.853, 0.852, 0.772, and 0.818 in the training, internal validation, and independent validation cohorts 1 and 2, respectively.

CONCLUSIONS

The fully automatic clinical-radiomics model based on deep learning and radiomics exhibits a good ability for hematoma segmentation and a favorable performance in stratifying HE risks.

A comprehensive prediction model predicts perihematomal edema growth in the acute stage after intracerebral hemorrhage

BACKGROUND

Perihematomal edema (PHE) is regarded as a potential intervention indicator of secondary injury following intracerebral hemorrhage (ICH). But it still lacks a comprehensive prediction model for early PHE formation.

METHODS

The included ICH patients have received an initial Computed Tomography scan within 6 hours of symptom onset. Hematoma volume and PHE volume were computed using semiautomated computer-assisted software. The volume of the hematoma, edema around the hematoma, and surface area of the hematoma were calculated. The platelet-to-lymphocyte ratio (PLR) was calculated by dividing the platelet count by the lymphocyte cell count. All analyses were 2-tailed, and the significance level was determined by P <0.05.

RESULTS

A total of 226 patients were included in the final analysis. The optimal cut-off values for PHE volume increase to predict poor outcomes were determined as 5.5 mL. For clinical applicability, we identified a value of 5.5 mL as the optimal threshold for early PHE growth. In the multivariate logistic regression analyses, we finally found that baseline hematoma surface area (p < 0.001), expansion-prone hematoma (p < 0.001), and PLR (p = 0.033) could independently predict PHE growth. The comprehensive prediction model demonstrated good performance in predicting PHE growth, with an area under the curve of 0.841, sensitivity of 0.807, and specificity of 0.732.

CONCLUSION

In this study, we found that baseline hematoma surface area, expansion-prone hematoma, and PLR were independently associated with PHE growth. Additionally, a risk nomogram model was established to predict the PHE growth in patients with ICH.

Perihaematomal Oedema Evolution over 2 Weeks after Spontaneous Intracerebral Haemorrhage and Association with Outcome: A Prospective Cohort Study

INTRODUCTION

We know little about the evolution of perihaematomal oedema (PHO) >24 h after ICH onset. We aimed to determine the trajectory of PHO after ICH onset and its association with outcome.

METHODS

We did a prospective cohort study using a pre-specified scanning protocol in adults with first-ever spontaneous ICH and measured absolute PHO volumes on CT head scans at ICH diagnosis and 3 ± 2, 7 ± 2, and 14 ± 2 days after ICH onset. We used the largest ICH if ICHs were multiple. The primary outcomes were (a) the trajectory of PHO after ICH onset and (b) the association between PHO (absolute volume at the time when most repeat CT head scans were obtained, and change in PHO volume at this time compared with the first CT head scan) and poor functional outcome (modified Rankin scale 3-6 at 90 days). We pre-specified multivariable logistic regression models of this association adjusting analyses for potential confounders: age, GCS, infratentorial ICH location, and intraventricular extension.

RESULTS

In 106 participants of whom 49 (46%) were female, with a median ICH volume 7 mL (interquartile range [IQR] 2-22 mL), the trajectory of median PHO volume increased from 14 mL (IQR: 7-26 mL) at diagnosis to 18 mL (IQR: 8-40 mL) at 3 ± 2 days (n = 87), 20 mL (IQR: 8-48 mL) at 7 ± 2 days (n = 93) and 21 mL (IQR: 10-54 mL) at 14 ± 2 days (n = 78) (p = <0.001). PHO volume at each time point was collinear with ICH volume at diagnosis (│r│ >0.7), but the change in PHO volume between diagnosis and each time point was not. Given collinearity, we used total lesion (i.e., ICH + PHO) volume instead of PHO volume in a logistic regression model of its association at each time point with outcome. Increasing total lesion (ICH + PHO) volume at day 7 ± 2 was associated with poor functional outcome (adjusted OR per mL 1.02, 95% CI: 1.00-1.03; p = 0.036), but the increase in PHO volume between diagnosis and day 7 ± 2 was not associated with poor functional outcome (adjusted OR per mL 1.03, 95% CI: 0.99-1.07; p = 0.132).

CONCLUSION

PHO volume increases throughout the first 2 weeks after onset of mild to moderate ICH. Total lesion (ICH + PHO) volume at day 7 ± 2 was associated with poor functional outcome, but the change in PHO volume between diagnosis and day 7 ± 2 was not. Prospective cohort studies with larger sample sizes are needed to investigate these associations and their modifiers.

Dynamic Imaging of Blood Coagulation Within the Hematoma of Patients With Acute Hemorrhagic Stroke

BACKGROUND

The dynamics of blood clot (combination of Hb [hemoglobin], fibrin, and a higher concentration of aggregated red blood cells) formation within the hematoma of an intracerebral hemorrhage is not well understood. A quantitative neuroimaging method of localized coagulated blood volume/distribution within the hematoma might improve clinical decision-making.

METHODS

The deoxyhemoglobin of aggregated red blood cells within extravasated blood exhibits a higher magnetic susceptibility due to unpaired heme iron electrons. We propose that coagulated blood, with higher aggregated red blood cell content, will exhibit (1) a higher positive susceptibility than noncoagulated blood and (2) increase in fibrin polymerization-restricted localized diffusion, which can be measured noninvasively using quantitative susceptibility mapping and diffusion tensor imaging. In this serial magnetic resonance imaging study, we enrolled 24 patients with acute intracerebral hemorrhage between October 2021 to May 2022 at a stroke center. Patients were 30 to 70 years of age and had a hematoma volume >15 cm3 and National Institutes of Health Stroke Scale score >1. The patients underwent imaging 3×: within 12 to 24 (T1), 36 to 48 (T2), and 60 to 72 (T3) hours of last seen well on a 3T magnetic resonance imaging system. Three-dimensional anatomic, multigradient echo and 2-dimensional diffusion tensor images were obtained. Hematoma and edema volumes were calculated, and the distribution of coagulation was measured by dynamic changes in the susceptibilities and fractional anisotropy within the hematoma.

RESULTS

Using a coagulated blood phantom, we demonstrated a linear relationship between the percentage coagulation and susceptibility (R2=0.91) with a positive red blood cell stain of the clot. The quantitative susceptibility maps showed a significant increase in hematoma susceptibility (T1, 0.29±0.04 parts per millions; T2, 0.36±0.04 parts per millions; T3, 0.45±0.04 parts per millions; P<0.0001). A concomitant increase in fractional anisotropy was also observed with time (T1, 0.40±0.02; T2, 0.45±0.02; T3, 0.47±0.02; P<0.05).

CONCLUSIONS

This quantitative neuroimaging study of coagulation within the hematoma has the potential to improve patient management, such as safe resumption of anticoagulants, the need for reversal agents, the administration of alteplase to resolve the clot, and the need for surgery.

Manifestations and Outcomes of Intracerebral Hemorrhage During the COVID-19 Pandemic in China: Multicenter, Longitudinal Cohort Study

BACKGROUND

The COVID-19 pandemic has inevitably affected the distribution of medical resources, and epidemic lockdowns have had a significant impact on the nursing and treatment of patients with other acute diseases, including intracerebral hemorrhage (ICH).

OBJECTIVE

This study aimed to investigate how the COVID-19 pandemic affected the manifestations and outcomes of patients with ICH.

METHODS

Patients with acute ICH before (December 1, 2018-November 30, 2019) and during (December 1, 2019-November 30, 2020) the COVID-19 pandemic at 31 centers in China from the Chinese Cerebral Hemorrhage: Mechanism and Intervention (CHEERY) study were entered into the analysis. Demographic information and clinical manifestations and outcomes were collected and compared between the 2 groups.

RESULTS

From December 1, 2018, to November 30, 2020, a total of 3460 patients with ICH from the CHEERY study were enrolled and eventually analyzed. The results showed that during the COVID-19 pandemic, patients with ICH were more likely to be older (P<.001) with a history of ischemic stroke (P=.04), shorter time from onset to admission (P<.001), higher blood pressure (P<.001), higher fasting blood glucose (P=.003), larger hematoma volume (P<.001), and more common deep ICH (P=.01) and intraventricular hemorrhage (P=.02). These patients required more intensive care unit treatment (P<.001) and preferred to go to the hospital directly rather than call an ambulance (P<.001). In addition, the COVID-19 pandemic was associated with an increased risk of pulmonary infection during hospitalization (adjusted risk ratio [RRadjusted] 1.267, 95% CI 1.065-1.509), lower probability of a 3-month good outcome (RRadjusted 0.975, 95% CI 0.956-0.995), and a higher probability of in-hospital (RRadjusted 3.103, 95% CI 2.156-4.465), 1-month (RRadjusted 1.064, 95% CI 1.042-1.087), and 3-month (RRadjusted 1.069, 95% CI 1.045-1.093) mortality.

CONCLUSIONS

Our study indicated that the cloud of COVID-19 has adversely impacted the presentation and outcomes of ICH. Medical workers may pay more attention to patients with ICH, while the public should pay more attention to hypertension control and ICH prevention.

TRIAL REGISTRATION

Chinese Clinical Trial Registry ChiCTR1900020872; https://www.chictr.org.cn/showprojEN.html?proj=33817.

Blood-Based Biomarkers in Intracerebral Hemorrhage

Intracerebral hemorrhage (ICH) is one of the most lethal subtypes of stroke, associated with high morbidity and mortality. Prevention of hematoma growth and perihematomal edema expansion are promising therapeutic targets currently under investigation. Despite recent improvements in the management of ICH, the ideal treatments are still to be determined. Early stratification and triage of ICH patients enable the adjustment of the standard of care in keeping with the personalized medicine principles. In recent years, research efforts have been concentrated on the development and validation of blood-based biomarkers. The benefit of looking for blood candidate markers is obvious because of their acceptance in terms of sample collection by the general population compared to any other body fluid. Given their ease of accessibility in clinical practice, blood-based biomarkers have been widely used as potential diagnostic, predictive, and prognostic markers. This review identifies some relevant and potentially promising blood biomarkers for ICH. These blood-based markers are summarized by their roles in clinical practice. Well-designed and large-scale studies are required to validate the use of all these biomarkers in the future.

Temporal Evolution of Perihematomal Blood-Brain Barrier Compromise and Edema Growth After Intracerebral Hemorrhage

PURPOSE

The aim of this study was to investigate the temporal evolution of perihematomal blood-brain barrier (BBB) compromise and edema growth and to determine the role of BBB compromise in edema growth.

METHODS

Spontaneous intracerebral hemorrhage patients who underwent computed tomography perfusion (CTP) were divided into five groups according to the time interval from symptom onset to CTP examination. Permeability-surface area product (PS) maps were generated using CTP source images. Ipsilateral and contralateral mean PS values were computed in the perihematomal and contralateral mirror regions. The relative PS (rPS) value was calculated as a ratio of ipsilateral to contralateral PS value. Hematoma and perihematomal edema volume were determined on non-contrast CT images.

RESULTS

In the total of 101 intracerebral hemorrhage patients, the ipsilateral mean PS value was significantly higher than that in contralateral region (z = -8.284, p < 0.001). The perihematomal BBB permeability showed a course of dynamic changes including an increase in the hyperacute and acute phases, a decrease in the early subacute phase and a second increase in the late subacute phase and chronic phase. Perihematomal edema increased gradually until the late subacute phase and then slightly increased. There was a relationship between rPS value and edema volume (β = 0.254, p = 0.006).

CONCLUSION

The perihematomal BBB permeability is dynamic changes, and edema growth is gradually increased in patients following intracerebral hemorrhage. BBB compromise plays an essential role in edema growth. The quantitative assessment of BBB compromise may provide valuable information in therapeutic interventions of intracerebral hemorrhage patients.

An updated systematic review and meta-analysis investigating perihematomal edema and clinical outcome after intracerebral hemorrhage

OBJECTIVES

The relationship between perihematomal edema (PHE) and intracerebral hemorrhage (ICH) outcomes is uncertain. Given newly published studies, we updated a previous systematic review and meta-analysis assessing the prognostic impact of PHE on ICH outcomes.

MATERIALS AND METHODS

Databases were searched through September 2022 using pre-defined keywords. Included studies used regression to examine the association between PHE and functional outcome (assessed by modified Rankin Scale [mRS]) and mortality. The study quality was assessed using the Newcastle-Ottawa Scale. The overall pooled effect, and secondary analyses exploring different subgroups were obtained by entering the log transformed odds ratios and their confidence intervals into a DerSimonian-Laird random effects meta-analysis.

RESULTS

Twenty-eight studies (n=8655) were included. The pooled effect size for overall outcome (mRS and mortality) was 1.05 (95% CI 1.03, 1.07; p<0.00). In secondary analyses, PHE volume and growth effect sizes were 1.03 (CI 1.01, 1.05) and 1.12 (CI 1.06, 1.19), respectively. Results of subgroup analyses assessing absolute PHE volume and growth at different time points were: baseline volume 1.02 (CI 0.98, 1.06), 72-hour volume 1.07 (CI 0.99, 1.16), growth at 24 hours 1.30 (CI 0.96, 1.74) and growth at 72 hours 1.10 (CI 1.04, 1.17). Heterogeneity across studies was substantial.

CONCLUSIONS

This meta-analysis indicates that PHE growth, especially within the first 24 hours after ictus, has a stronger impact on functional outcome and mortality than PHE volume. Definitive conclusions are limited by the large variability of PHE measures, heterogeneity, and different evaluation time points between studies.

Prediction of Early Perihematomal Edema Expansion Based on Noncontrast Computed Tomography Radiomics and Machine Learning in Intracerebral Hemorrhage

OBJECTIVES

To investigate the predictive value of noncontrast computed tomography (NCCT) models based on radiomics features and machine learning for early perihematomal edema (PHE) expansion in patients with spontaneous intracerebral hemorrhage (ICH).

METHODS

We retrospectively reviewed NCCT data from 214 patients with spontaneous ICH. All radiomics features were extracted from volume of interest of hematomas on admission scans. A total of 8 machine learning methods were applied for constructing models in the training and the test set. Receiver operating characteristic analysis and the areas under the curve were used to evaluate the predictive value.

RESULTS

A total of 23 features were finally selected to establish models of early PHE expansion after feature screening. Patients were randomly assigned into training (n = 171) and test (n = 43) sets. The accuracy, sensitivity, and specificity in the test set were 72.1%, 90.0%, and 66.7% for the support vector machine model; 79.1%, 70.0%, and 84.4% for the k-nearest neighbor model; 88.4%, 90.0%, and 87.9% for the logistic regression model; 74.4%, 90.0%, and 69.7% for the extra tree model; 74.4%, 90.0%, and 69.7% for the extreme gradient boosting model; 83.7%, 100%, and 78.8% for the multilayer perceptron (MLP) model; 72.1%, 100%, and 65.6% for the light gradient boosting machine model; and 60.5%, 90.0%, and 53.1% for the random forest model, respectively.

CONCLUSIONS

The MLP model seemed to be the best model for prediction of PHE expansion in patients with ICH. NCCT models based on radiomics features and machine learning could predict early PHE expansion and improve the discrimination of identify spontaneous intracerebral hemorrhage patients at risk of early PHE expansion.

Neutrophil percentage to albumin ratio is associated with stroke-associated pneumonia and poor outcome in patients with spontaneous intracerebral hemorrhage

Background

The purpose of this study was to investigate the diagnostic performance of the neutrophil percentage-to-albumin ratio (NPAR) for predicting stroke-associated pneumonia (SAP) and functional outcome in patients with intracerebral hemorrhage (ICH).

Methods

We analyzed our prospective database of consecutive ICH patients who were admitted to the First Affiliated Hospital of Chongqing Medical University from January 2016 to September 2021. We included subjects with a baseline computed tomography available and a complete NPAR count performed within 6h of onset. The patients' demographic and radiological characteristics were analyzed. Good outcome was defined as a modifed Rankin Scale score of 0-3 at 90 days. Poor outcome was defined as a modifed Rankin Scale score of 4-6 at 90 days. Multivariable logistic regression models were used to investigate the association between NPAR, SAP, and functional outcome. Receiver operating characteristic (ROC) curve analysis was conducted to identify the optimal cutoff of NPAR to discriminate between good and poor outcomes in ICH patients.

Results

A total of 918 patients with ICH confirmed by non-contrast computed tomography were included. Of those, 316 (34.4%) had SAP, and 258 (28.1%) had poor outcomes. Multivariate regression analysis showed that higher NPAR on admission was an independent predictor of SAP (adjusted odds ratio: 2.45; 95% confidence interval, 1.56-3.84; P<0.001) and was associated with increased risk of poor outcome (adjusted odd ratio:1.72; 95% confidence interval, 1.03-2.90; P=0.040) in patients with ICH. In ROC analysis, an NPAR of 2 was identified as the optimal cutoff value to discriminate between good and poor functional outcomes.

Conclusion

Higher NPAR is independently associated with SAP and poor functional outcome in patients with ICH. Our findings suggest that early prediction of SAP is feasible by using a simple biomarker NPAR.

The association between perihaematomal oedema and functional outcome after spontaneous intracerebral haemorrhage: A systematic review and meta-analysis

PURPOSE

Perihaematomal oedema (PHO) formation has gained increasing interest as a therapeutic target after spontaneous intracerebral haemorrhage (ICH). Whether PHO contributes to poor outcome is unclear. We aimed to determine the association between PHO and outcome in patients with spontaneous ICH.

METHOD

We searched five databases up to 17 November 2021 for studies of ⩾10 adults with ICH reporting the presence of PHO and outcome. We assessed risk of bias, extracted aggregate data and used random effects meta-analysis to pool studies that reported odds ratios (OR) with 95% confidence intervals (CI). Primary outcome was poor functional outcome defined as modified Rankin Scale score of 3-6 at 3 months. Additionally, we assessed PHO growth and poor outcome at any time of follow-up. We prospectively registered the protocol in PROSPERO (CRD42020157088).

FINDINGS

We identified 12,968 articles, of which we included 27 studies (n = 9534). Eighteen studies reported an association between larger PHO volume and poor outcome, six a neutral result and three an inverse relationship. Larger absolute PHO volume was associated with poor functional outcome at 3 months (OR per mL increase of absolute PHO 1.03, 95% CI 1.00-1.06, I2 44%, four studies). Additionally, PHO growth was associated with poor outcome (OR 1.04, 95% CI 1.02-1.06, I2 0%, seven studies).

DISCUSSION

In patients with spontaneous ICH, larger PHO volume is associated with poor functional outcome at 3 months. These findings support the development and investigation of new therapeutic interventions targeting PHO formation to evaluate if reduction of PHO improves outcome after ICH.

Comparison of different noncontrast computed tomographic markers for predicting early perihematomal edema expansion in patients with intracerebral hemorrhage

OBJECTIVES

Noncontrast computed tomography (NCCT) imaging markers are associated with early perihematomal edema (PHE) growth. The aim of this study was to compare the predictive value of different NCCT markers in predicting early PHE expansion.

METHODS

ICH patients who underwent baseline CT scan within 6 h of symptoms onset and follow-up CT scan within 36 h between July 2011 and March 2017 were included in this study. The predictive value of hypodensity, satellite sign, heterogeneous density, irregular shape, blend sign, black hole sign, island sign and expansion-prone hematoma for early perihematomal edema expansion were assessed, separately.

RESULTS

214 patients were included in our final analysis. After adjusting for ICH characteristics, hypodensity, blend sign, island sign and expansion-prone hematoma are still predictors of early perihematomal edema expansion in multivariable logistics regression analysis (all P < 0.05). The area under the receiver operating characteristic (ROC) curve of expansion-prone hematoma was significantly larger than the area under the ROC curve of hypodensity, blend sign and island sign in predicting PHE expansion (P = 0.003, P < 0.001 and P = 0.002, respectively).

CONCLUSION

Compared with single NCCT imaging markers, expansion-prone hematoma seems to be optimal predictor for early PHE expansion than any single NCCT imaging marker.

Off-Hour Admission Is Associated with Poor Outcome in Patients with Intracerebral Hemorrhage

The mortality of stroke increases on weekends and during off-hour periods. We investigated the effect of off-hour admission on the outcomes of intracerebral hemorrhage (ICH) patients. We retrospectively analyzed a prospective cohort of ICH patients, admitted between January 2017 and December 2019 at the First Affiliated Hospital of Chongqing Medical University. Acute ICH within 72 h after onset with a baseline computed tomography and 3-month follow-up were included in our study. Multivariable logistic regression analysis was performed for calculating the odds ratios (OR) and 95% confidence interval (CI) for the outcome measurements. Of the 656 participants, 318 (48.5%) were admitted during on-hours, whereas 338 (51.5%) were admitted during off-hours. Patients with a poor outcome had a larger median baseline hematoma volume, of 27 mL (interquartile range 11.1-53.2 mL), and a lower median time from onset to imaging, of 2.8 h (interquartile range 1.4-9.6 h). Off-hour admission was significantly associated with a poor functional outcome at 3 months, after adjusting for cofounders (adjusted OR 2.17, 95% CI 1.35-3.47; p = 0.001). We found that patients admitted during off-hours had a higher risk of poor functional outcomes at 3 months than those admitted during working hours.

Molecular, Pathological, Clinical, and Therapeutic Aspects of Perihematomal Edema in Different Stages of Intracerebral Hemorrhage

Acute intracerebral hemorrhage (ICH) is a devastating type of stroke worldwide. Neuronal destruction involved in the brain damage process caused by ICH includes a primary injury formed by the mass effect of the hematoma and a secondary injury induced by the degradation products of a blood clot. Additionally, factors in the coagulation cascade and complement activation process also contribute to secondary brain injury by promoting the disruption of the blood-brain barrier and neuronal cell degeneration by enhancing the inflammatory response, oxidative stress, etc. Although treatment options for direct damage are limited, various strategies have been proposed to treat secondary injury post-ICH. Perihematomal edema (PHE) is a potential surrogate marker for secondary injury and may contribute to poor outcomes after ICH. Therefore, it is essential to investigate the underlying pathological mechanism, evolution, and potential therapeutic strategies to treat PHE. Here, we review the pathophysiology and imaging characteristics of PHE at different stages after acute ICH. As illustrated in preclinical and clinical studies, we discussed the merits and limitations of varying PHE quantification protocols, including absolute PHE volume, relative PHE volume, and extension distance calculated with images and other techniques. Importantly, this review summarizes the factors that affect PHE by focusing on traditional variables, the cerebral venous drainage system, and the brain lymphatic drainage system. Finally, to facilitate translational research, we analyze why the relationship between PHE and the functional outcome of ICH is currently controversial. We also emphasize promising therapeutic approaches that modulate multiple targets to alleviate PHE and promote neurologic recovery after acute ICH.

Defining Delayed Perihematomal Edema Expansion in Intracerebral Hemorrhage: Segmentation, Time Course, Risk Factors and Clinical Outcome

We attempt to generate a definition of delayed perihematomal edema expansion (DPE) and analyze its time course, risk factors, and clinical outcomes. A multi-cohort data was derived from the Chinese Intracranial Hemorrhage Image Database (CICHID). A non-contrast computed tomography (NCCT) -based deep learning model was constructed for fully automated segmentation hematoma and perihematomal edema (PHE). Time course of hematoma and PHE evolution correlated to initial hematoma volume was volumetrically assessed. Predictive values for DPE were calculated through receiver operating characteristic curve analysis and were tested in an independent cohort. Logistic regression analysis was utilized to identify risk factors for DPE formation and poor outcomes. The test cohort’s Dice scores of lesion segmentation were 0.877 and 0.642 for hematoma and PHE, respectively. Overall, 1201 patients were enrolled for time-course analysis of ICH evolution. A total of 312 patients were further selected for DPE analysis. Time course analysis showed the growth peak of PHE approximately concentrates in 14 days after onset. The best cutoff for DPE to predict poor outcome was 3.34 mL of absolute PHE expansion from 4-7 days to 8-14 days (AUC=0.784, sensitivity=72.2%, specificity=81.2%), and 3.78 mL of absolute PHE expansion from 8-14 days to 15-21 days (AUC=0.682, sensitivity=59.3%, specificity=92.1%) in the derivation sample. Patients with DPE was associated with worse outcome (OR: 12.340, 95%CI: 6.378-23.873, P<0.01), and the larger initial hematoma volume (OR: 1.021, 95%CI: 1.000-1.043, P=0.049) was the significant risk factor for DPE formation. This study constructed a well-performance deep learning model for automatic segmentations of hematoma and PHE. A new definition of DPE was generated and is confirmed to be related to poor outcomes in ICH. Patients with larger initial hematoma volume have a higher risk of developing DPE formation.